Date   : Mon, 18 Jul 1994 10:23:24 EST
From   : Stephen Quan <quan@...>
Subject: Re: My progress so far...

Chris Rae writes

> I've got back to work on my 6502 emulator now... I've managed to speed it 
> up a bit. Here's where assembler beats C any day. ;-)
>
> .... but I can now 
> execute the three-nested loop we've been going on about in 50.04 seconds 
> on my 386/40; if anyone would like to test this against the C version, 
> just for interest, I'll post the EXE. The source is just over 600 lines 
> now...

Ok, I am prepared to front up my emulator!  BTW, what compiler are you
using?  I just switched over to using DJGPP compiler for DOS, and the
compiled code it generates was far faster than my previous Borland C
version.  I am getting that 256^3 nested loop to run at

    59 seconds on a 386DX-40
    31 seconds on a 486DX-33
    23 seconds on a Sparc-2
    15 seconds on a 486DX2-66

I have not seen the original thread and I might not be using the same
256^3 assembly fragment as everyone else is using.  I have posted the
one that I am using, but I fear it may not have made it out to the list.
This same loop when tried on my BBC-B ran at 44 seconds, so on all
platforms except the 386DX-40, I am running faster.
-- 
Stephen Quan (quan@...                 ), SysAdmin, Analyst/Programmer.
Centre for Spatial Information Studies, University of Tasmania, Hobart.
GPO BOX 252C, Australia, 7001.  Local Tel: (002) 202898 Fax: (002) 240282
International Callers use +6102 instead of (002).